Bacterial oxidation of ferrous iron at low temperatures

ARTICLE

Bacterial Oxidation of Ferrous Iron atLow Temperatures

Daniel Kupka,1 Olena I. Rzhepishevska,2 Mark Dopson,2 E. Borje Lindstrom,2

Olia V. Karnachuk,3 Olli H. Tuovinen4

1Institute of Geotechnics, Slovak Academy of Sciences, Watsonova 45,

SK-043 53 Kosice, Slovakia2Department of Molecular Biology, Umea University, SE-901 87 Umea, Sweden;

telephone: þ46 (0) 90 7856769; fax: þ46 (0) 90 772630;

e-mail: [email protected] of Agriculture and Environmental Science, Tomsk State University,

Prospekt Lenina 36, 634050 Tomsk, Russia4Department of Microbiology, Ohio State University, 484 West 12th Avenue,

Columbus, Ohio 43210

Received 13 December 2006; accepted 20 January 2007

Published online 15 February 2007 in Wiley InterScience (www.interscience.wiley.com
). DOI 10.1002/bit.21371
ABSTRACT: This study comprises the first report offerrous iron oxidation by psychrotolerant, acidophiliciron-oxidizing bacteria capable of growing at 58C. Samplesof mine drainage-impacted surface soils and sediments fromthe Norilsk mining region (Taimyr, Siberia) and Kristine-berg (Skellefte district, Sweden) were inoculated into acidicferrous sulfate media and incubated at 58C. Iron oxidationwas preceded by an approximately 3-month lag period thatwas reduced in subsequent cultures. Three enrichmentcultures were chosen for further work and one culturedesignated as isolate SS3 was purified by colony isolationfrom a Norilsk enrichment culture for determining thekinetics of iron oxidation. The 16S rRNA based phylogenyof SS3 and two other psychrotolerant cultures, SS5 fromNorilsk and SK5 from Northern Sweden, was determined.Comparative analysis of amplified 16S rRNA gene sequencesshowed that the psychrotolerant cultures aligned withinAcidithiobacillus ferrooxidans. The rate constant of ironoxidation by growing cultures of SS3 was in the range of0.0162–0.0104 h�1 depending on the initial pH. The oxida-tion kinetics followed an exponential pattern, consistentwith a first order rate expression. Parallel iron oxidationby a mesophilic reference culture of Acidithiobacillusferrooxidans was extremely slow and linear. Precipitatesharvested from the 58C culture were identified by X-raydiffraction as mixtures of schwertmannite (ideal formulaFe8O8(OH)6SO4) and jarosite (KFe3(SO4)2(OH)6). Jarositewas much more dominant in precipitates produced at 308C.Biotechnol. Bioeng. 2007;97: 1470–1478.

� 2007 Wiley Periodicals, Inc.

Correspondence to: M. Dopson

1470 Biotechnology and Bioengineering, Vol. 97, No. 6, August 15, 2007

KEYWORDS: Acidithiobacillus ferrooxidans; iron oxidation;jarosite; molecular phylogeny; psychrotolerant; schwert-mannite

Introduction

Iron-oxidizing micro-organisms such as Acidithiobacillusferrooxidans are widely found in environments impacted bymining and mine drainage and constitute a phylogeneticallyheterogeneous group among the Bacteria and Archaea(Hallberg and Johnson, 2001; Norris, 2007). These micro-organisms oxidize a wide variety of sulfide minerals,releasing iron (and other metals) and sulfo-oxyanions inthe solution and contribute to acid mine drainage problems(Rohwerder et al., 2003). Acidophilic iron-oxidizing micro-organisms that are similar to indigenous organisms in minesites are employed in biological-leaching applications for theremoval of metals from sulfide ores (Olson et al., 2003;Rohwerder et al., 2003). Iron oxidation is initially acid-consuming:

4Fe2þ þ O2 þ 4Hþ ! 4Fe3þ þ 2H2O ð1Þ

Subsequent reactions of iron hydrolysis, leading toschwertmannite (Eq. 2) and jarosite (Eqs. 3 and 4)formation are acid producing and the final pH value

� 2007 Wiley Periodicals, Inc.

depends on the extent of equilibration and type of ferric ironprecipitation.

8Fe3þ þ SO2�4 þ 14H2O ! Fe8O8ðOHÞ6SO4 þ 22Hþ (2)

3Fe3þ þ Kþ þ 2SO2�4 þ 6H2O

! KFe3ðSO4Þ2ðOHÞ6 þ 6Hþ (3)

Fe8O8ðOHÞ6SO4 þ SO2�4 þ Kþ þ 16Hþ

! KFe3ðSO4Þ2ðOHÞ6 þ 5Fe3þ þ 8H2O (4)

Biological oxidation of iron and sulfide minerals occursover a wide range of climatic conditions and numerousreports describe acidophilic, mesophilic, and thermophiliciron-oxidizing isolates (Norris, 2007; Plumb et al., 2007;Rawlings, 2002; Rohwerder et al., 2003). Mesophilic ironoxidizers have been isolated from environments with extremeannual temperature variations, but their activity is con-siderably lower at temperatures approaching 58C. Bacterialoxidation of iron and pyrite (FeS2) in low temperatureenvironments has been previously reported (Elberling, 2003;Elberling et al., 2000; Langdahl and Ingvorsen, 1997), butthese processes are generally poorly characterized in arcticenvironments. Ferroni et al. (1986) enriched for iron-oxidizing cultures at six incubation temperatures usingsamples from a closed uranium mine in the Elliot Lake area,Ontario, Canada. The generation times determined for theenrichment cultures at 2 and 68C were 247� 46 and 103�29 h, respectively, compared to 12 h in the mesophilic rangeof 25–308C. Berthelot et al. (1993) extended this work byadditional enrichments and isolations of At. ferrooxidansfrom two closed uraniummine sites. They reported that mostisolates (91%) grew at 48C as well as at mesophilictemperatures, but all isolates grew fastest at temperaturesover 218C. Similarly, Leduc et al. (1993) reported on severalstrains that were capable of growth at 28C. All these resultssuggest that the isolates are psychrotolerant (optimumgrowth at 20–408C but can grow at 2–48C), but notpsychrophilic (optimum growth at�158C and lack of growthat 218C). Ahonen and Tuovinen (1989) enriched for Fe2þ-oxidizers at temperatures as low as 48C and the iron oxidationrates measured with those cultures were used to constructan Arrhenius plot that yielded an activation energy (Ea) of83 kJ�mol�1. The effect of temperature on the kinetics ofgrowth and substrate oxidation by iron-oxidizing bacteria hasbeen addressed in numerous studies (Ahonen et al., 1990;Ahonen and Tuovinen, 1990; Ahonen and Tuovinen, 1989;Franzmann et al., 2005; Kovalenko et al., 1981; Leduc et al.,1993; Okereke and Stevens, 1991). In general, decreasedtemperature reduces biological activity and the effect ofincreasing temperature can be describedwith temperature co-efficient (Q10) values of approximately 2.

The objective of this study was to enrich for and isolateiron-oxidizing bacteria that grow faster at 58C than

mesophiles using samples from an arctic site impacted bythe mining industry. Oxidation of ferrous iron and growthof iron-oxidizing micro-organisms is not only of ecologicaland biotechnological interest, but it also interlinks withthe kinetics and solid-phase products of biogeochemicaltransformations of iron in arctic and other borealenvironments. Since temperature is known to affect theprecipitation of Fe(III)-hydroxysulfates (Wang et al.,2007), the products of iron precipitation at two differenttemperatures were also characterized in this study.

Materials and Methods

Sampling and Enrichment

Sampling was performed in September 2002 in the Norilskarea, Taimyr Peninsula, Northern Siberia (300 km north ofthe Arctic Circle). The region is a permafrost area andfrom 1934–1994 the mean temperature was �9.68C with aminimum of�53.18C (Savchenko and Novitsky, 2001). Theaverage depth of the permafrost table is 2 m and is capped bya surface layer, which is subject to thawing during the warmseasons. The average temperature of the top soil is <08C inOctober through May with a minimum of �27.98C, anaverage July temperature of 16.18C, and maximum recordedtemperature of 19.08C (Savchenko and Novitsky 2001).Although the average summer temperatures reach 16.18C,frosts may still occur at night. The area contains Ni–Cusulfide ores worked in open cast and underground mineswith associated mineral processing and tailings areas. Thefastest growing enrichment culture was SS3 (site T2) fromsurface layers (0–5 cm) of Fe(III)-containing sedimentmixed with waste rock and inundated by seepage fromLebiazie ore mill tailings dam (Karnachuk et al., 2005).Sample SS5 was a composite of several subsamples collectedfrom different mine waste-impacted areas of the Norilskindustrial complex.

Sampling was also performed in October 2002 in thetailings impoundment area in Kristineberg, NorthernSweden. The geochemical characteristics of the site havebeen previously described (Holmstrom et al., 2001). TheKristineberg mine is in the Skellefte ore district, containinga massive Zn–Cu ore body. The average annual airtemperature at the site is 18C with 5 months of averagetemperature<08C. The site was used in the 1940s and 1950sfor deposition of tailings from the enrichment plant atKristineberg, which also processed ore from other minescloseby. Sample SK5 was collected from oxidized surfacedrainage and precipitates at the site.

In the laboratory, samples SS3, SS5, and SK5 wereinoculated into ferrous sulfate media that contained(per liter) 0.5 g each of (NH4)2SO4, K2HPO4, andMgSO4�7H2O, and 33.4 g FeSO4�7H2O, initial pH 2.2.The medium was prepared in two stock solutions, of whichthe ferrous sulfate solution was filter-sterilized and mineralsalts solution was autoclaved. The enrichment cultures were

Kupka et al.: Iron Oxidation at 58C 1471

Biotechnology and Bioengineering. DOI 10.1002/bit

incubated in shake flasks at 5� 18C. Enrichment culturesfrom this initial screening were subcultured twice with 10%inocula in ferrous sulfate media at 58C and sparged withfilter sterilized air during the incubation.

Growth Experiments

A pure culture of At. ferrooxidans (designated as strain SS3)was isolated from the SS3 enrichment culture for growthexperiments. An iron-oxidizing isolate of At. ferrooxidans froman acid mine drainage site in Smolnik (East Slovakia) was usedas a mesophilic reference culture (Kupka 2004). For growthexperiments, the pure SS3 culture and the Smolnik isolate weregrown in 9K medium at 5� 18C in a refrigerated incubator.The medium contained (g L�1) 3 g (NH4)2SO4, 0.1 g KCl,0.5 g K2HPO4, 0.5 g MgSO4�7H2O, 0.01 g Ca(NO3)2, and44.2 g FeSO4�7H2O (Silverman and Lundgren 1959), pHadjusted with H2SO4 as required, and autoclaved. Filtersterilized ferrous sulfate solution was added before inoculation(3% (vol/vol)). For plating, themediumwas solidified with 1%(wt/vol) agarose (agarose I, Amresco, Solon, Ohio). Growthexperiments including abiotic controls were performed in150 mL cultures in 250 mL flasks stirred at 320 rpm. Forcomparison of growth and ferrous iron oxidation rates, themesophilic Smolnik strain was grown at 308C (as describedabove). Samples were removed for pHmeasurements and ironanalysis. Suspended solids were separated by centrifugation(12,000� g for 15 min). Ferrous iron concentration wasdetermined by titrimetric method with K2Cr2O7 with adiphenylamine as the indicator (Vogel, 1961). Ferric iron wasdetermined by a UV-spectrophotometric method (Barasanand Tuovinen, 1986) and total soluble iron by the sum of Fe2þ

and Fe3þ and by atomic absorption spectroscopy (Spectra AA-30, Varian Scientific Instruments, Mulgrave, Vic., Australia).Data for At. ferrooxidans SS3 and the reference strain areaverages (number of replicates (n)¼ 2)� SD. For growthexperiments with tetrathionate (K2S4O6) or organic carboncompounds aminimal salts medium containing trace elements(pH 2.5) was used (Dopson and Lindstrom, 1999). Cultureswere incubated in shake flasks (200 rpm) with tetrathionateat 5 and 308C and with organic compounds at 308C. Growthwas measured by increase in optical density at 600 nm(OD600). With the exception of cas amino acids (2% (wt/vol))the organic compounds were tested at 0.02% (wt/vol). Thecompounds were yeast extract, cas amino acids, and glucosefor mixotrophic and organotrophic growth; and sucrose,glycerol, arabinose, fructose, galactose, and xylose foronly mixotrophic growth. The data are presented as means(n¼ 2–3)� SD.

Measurement of Rates of Ferrous IronOxidation by Resting Cells

Bacteria were harvested from exponentially growing culturesby filtration (0.45 mm pore size membrane filter). Cells werewashed three times with acidified water (pH 2.0) and


suspended into fresh media adjusted to pH 2.0. Ironoxidation rates were measured under substrate-excessconditions in a jacketed reaction cell maintained at aconstant temperature with a circulating water bath. Thereaction mixture (30 mL) was stirred at 400 rpm with amagnetic stirrer. The oxygen uptake rate was measured witha polarographic (Clark-type) oxygen electrode in a closedsystem. Reaction rates were calculated from steady linearoxygen uptake rates obtained at the temperature range of5–308C. Results are presented as averages (n¼ 3)� SD.

X-Ray Diffraction

Precipitates were harvested from spent cultures bymembrane (0.45 mm) filtration, washed three times withdistilled water followed by air drying at 22� 28C. Powderedsamples were analyzed by X-ray diffraction (CuKa radia-tion) using a Philips PW1070 goniometer equipped with adiffracted-beam monochrometer and a theta-compensatingslit. Samples were scanned from 3 to 808 2u with a stepincrement of 0.058 2u and 4 s counting time.

Analysis of 16S rRNA Genes

Cells (100 mL culture) for DNA extraction and PCRamplification were pelleted (10,000� g for 10 min), washedtwice with 10 mM Tris pH 8, and re-suspended in 0.5 mL10 mM Tris HCl, 1 mM EDTA buffer containing 20 mg/mLlysozyme, and incubated at 378C for 30 min. Proteinase K(200 mg) was added and the cells further incubated at 608Cfor 5 h, followed by inactivation at 708C for 30 min. Cellswere sheared by vortexing with glass beads (425–600 mm;Sigma, Stockholm, Sweden) and DNA was isolated usingthe Wizard DNA Clean Up System (Promega, Sundbyberg,Sweden) (Dopson and Lindstrom, 2004). To test thephylogeny of SS3, SS5, and SK5, approximately 550 base pairfragments of the 16S rRNA genes were amplified utilizingPCR primers DS907R and GM5F with a GC clamp ((Muyzeret al., 1995); Table 1) and separated by denaturing gradientgel electrophoresis (DGGE) with a denaturing gradient of35–60% (Dopson and Lindstrom, 2004).

For sequencing of the 16S rRNA gene, DNA was PCRamplified from all three cultures utilizing primer pairs 27Fand DS907R and 533F and 1492R (Table 1), cloned usingthe pGEM-T Easy Vector System (Promega, Falkenberg,Sweden), and transformed into Escherichia coli strain DH5a(laboratory stock) (Dopson and Lindstrom, 2004). Thecloned 16S rRNA genes were PCR amplified in preparationfor sequencing in both directions using the DYEnamic ETTerminator Cycle Sequencing Premix Kit (AmershamPharmacia, Uppsala, Sweden) and oligonucleotides M13universal and M13 reverse (Amersham Pharmacia). ThePCR products were sequenced using an ABI Prism 377 DNASequencer.

DOI 10.1002/bit

Table I. Oligonucleotides utilized in this study.

Primer Sequences (50-30) Reference

GM5F-GCa CCTACGGGAGGCAGCAG Muyzer et al. (1995)

DS907R CCG TCA ATT CCT TTR AGT TT Muyzer et al. (1995)

27F AGAGTTTGATCCTGGCTCAG Lane (1985)

533F GTGCCAGCCGCCGCGGTAAþGTGCCAGCAGCCGCGGTAA Lane (1985)

1492R GGTTACCTTGTTACGACTT Lane (1985)

M13F GTAAAACGACGGCCAGT Messing (1983)

M13R CAGGAAACAGCTATGACCATG Messing (1983)

aGC-clamp (50-CGCCCGCCGCGCCCCGCGCCCGTCCCGCCGCCCCCGCCCG-30) is attached to the 50 endof the primer (Muyzer and Smalla 1998).

Phylogenetic Analysis

The lengths of the 16S rRNA gene sequences obtained were1497, 1505, and 1507 base pairs for strains SS3, SK5, andSS5, respectively. Evolutionary analyses of 16S rRNA genesequences were performed by alignment using ARB (Strunkand Ludwig, 2002) that was then used to create a distancematrix using a Jukes and Cantor correction. Three separatephylogenetic trees were created by DNA distance andneighbor joining, DNA parsimony, and maximum like-lihood methods using ARB. The neighbour joiningtree thus constructed includes the nodes supported by thedifferent trees. Accession numbers for the SS3, SK5, andSS5 16S rRNA gene sequences are AY960976, AY960977, andAY960978, respectively.

Results and Discussion

Enrichment Cultures and Purification

Initial enrichment cultures with Norilsk samples tookseveral months to show signs of ferrous iron oxidation whenincubated in shake flasks at 5� 18C. The most rapid ironoxidation took place in Norilsk enrichment cultures SS3 andSS5 and the Kristineberg SK5 enrichment culture and theywere selected for further experiments. For culture purifica-tion, the enrichment culture SS3 was plated on ferroussulfate agarose medium and a single colony isolated forsubsequent liquid culture experiments. The cells wereGram-negative single and paired rods and highly motile.The isolate SS3 and the enrichment cultures SS5 and SK5were further characterized by analysis of 16S rRNA genes.

Phylogenetic Characterization by Analysis of16S rRNA Genes

16S rRNA genes isolated from SS3, SS5, and SK5 gave singlebands on the DGGE gel (Fig. 1A), suggesting that these threecultures did not contain multiple dominant phylotypes.The 16S rRNA gene sequences of cultures SS3, SS5, and SK5were sequenced and used to construct a phylogenetictree (Fig. 1B). The three cultures formed a clade withAt. ferrooxidans and were 98–99% similar to an uncultured

bacterium clone RA13C4 isolated from an in situreactor system treating monochlorobenzene contaminatedgroundwater (Alfreider et al., 2002). At. ferrooxidans KSC1,NO-8, �25, and �37 were all identified from King’s Mine(Kongsberg, central Norway), an abandoned copper minethat typically has <08C temperatures for 7 months ofthe year (Johnson et al., 2001). Karavaiko et al. (2003)subsequently confirmed their phylogenetic assignmentto the species At. ferrooxidans. The closely relatedAt. ferrooxidans T7 was recently identified (Dopson et al.,2007) from an enrichment culture derived from abioleaching mine site in Finland and used for columnleaching of a black schist (Puhakka et al., 2007) at 78C.

Growth of At. ferrooxidans SS3 on Ferrous Iron,Tetrathionate, and Organic Carbon

After a lag period of approximately 1 month, isolate SS3oxidized iron rapidly at an initial pH of 2.0 but not at pH 1.6.The mesophilic reference strain had a linear, slow ironoxidation rate at 58C and iron oxidation was negligible inthe chemical controls (Fig. 2A). Further growth experimentswere carried out in the initial pH range of 2.0 and 2.8 duringwhich the lag period was reduced from 600 h toapproximately 1 week (Fig. 2B). Culture SS3 oxidized ironfast at initial pH 2.8 at 58C. The rates were linear on semi-logplots (Fig. 2C), in agreement with the first-order rateexpression of iron oxidation. Because the rate of ironoxidation was linear on a semi-log plot (Fig. 2), thecorresponding rate constants were calculated from therespective slopes. The rate constants (Table 2) ranged from0.016 to 0.012 h�1, which is equivalent to a generation timeof 43–58 h. Iron oxidation by At. ferrooxidans SS3 wasmeasured at an initial pH range of 1.8–2.8 (Fig. 3). Data inFigure 3 also include the changes in soluble ferric iron andsoluble total iron concentration and show iron precipitationduring the latter half of the time course. The concentrationof soluble ferric iron decreased towards the end of thetime course (Fig. 3). The pH values increased upon ironoxidation at the start of the incubations and then decreasedupon subsequent ferric iron precipitation and equilibration(Fig. 4). The pH values of cultures grown at initial pH 1.8–2.2 first increased by approximately 0.5 pH units (reaching a



Figure 1. A: DGGE gel of partial 16S rRNA genes amplified from SS3, SS5, and SK5. B: Neighbor joining tree based on 16S rRNA gene sequences isolated in this study (in bold)

and sequences from the database. Phylogenetic analysis was carried out by the maximum likelihood, distance neighbor joining, and DNA parsimony methods in ARB and the nodes

supported by all three trees (&) and two trees (&) have been marked. Accession numbers are given in parenthesis. The scale bar corresponds to 10% sequence similarity.

peak between after approximately 160–310 h) andthen decreased. The pH values of cultures at an initialpH of 2.5 and 2.8 only increased slightly and decreased afterapproximately 120 h. At low temperature, much more of theferric iron remained in solution despite the pH rising above1.6. The increase is attributed to proton consumptionassociated with iron oxidation (Eq. 1) and acid production isattributed to schwertmannite or jarosite precipitation(Eqs. 2–4). Dopson et al. (2007) showed that the increasedsolubility of ferric iron at low temperature helped maintaincomparable column leaching rates at 7 and 218C. As a pointof interest for pH control, slow precipitation of ferriciron at low temperature reduces the decrease in the pHthat accompanies Fe(III)-oxyhydroxysulfate precipitation(Eqs. 2 and 3). Thus, the pH and ferric iron concentrationeffects may also influence the rates of bacterial oxidation offerrous iron. Such interactive effects are important tounderstand when bioleaching processes are operated withparameters that are outside the cardinal values.

At. ferrooxidans SS3 was grown with 5 mM tetrathionateat 5 and 308C as sole energy source. Growth on tetrathionatestarted after a lag time of 13 and 8 days and reachedan OD600 of 0.110� 0.021 (n¼ 2) and 0.092� 0.007 (n¼ 3)for 5 and 308C cultures, respectively (Fig. 5).None of theorganic carbon sources supported organotrophic growth ofAt. ferrooxidans SS3. Their presence in tetrathionate mediadid not accelerate growth or increase the yield in


comparison to chemolithoautotrophic growth on tetra-thionate alone.

Effect of Temperature on Iron Oxidation by CellSuspensions of At. ferrooxidans SS3

Iron oxidation rates at different temperatures werecalculated from the corresponding oxygen uptake measure-ments in accordance with Eq. 1. Rates obtained at 5–308Cwere used to calculate the activation energy by linearizationof the Arrhenius equation. A linear relationship wasobtained when the ln k values were fitted against thereciprocal absolute temperatures in the range of 5–258C(Fig. 6). The value determined for 308C incubation wasexcluded from calculation of the Ea because it appeared todeviate from the linearity of the regression line. Therefore,the optimum temperature for ferrous iron oxidation wasbetween 25 and 308C, and some decline in the rate ofbiological iron oxidation was already apparent at 308C.

The Ea value was 38 kJ�mol�1 calculated from the oxygenuptake rate measurements in the 5–258C range in this work.This value is very similar to the Ea of 36 kJ mol�1 calculatedfrom rate measurements of continuous iron oxidation in therange of 30–408C for a mixed culture dominated byLeptospirillum ferrooxidans (Breed et al., 1999). Nemati andWebb (1997) reported an Ea value of 68 kJ mol�1, calculated

DOI 10.1002/bit

Figure 2. Iron oxidation by At. ferrooxidans SS3 pure culture at 58C. (A) Ironoxidation rates by At. ferrooxidans SS3 at an initial medium pH of 2.0 (") and pH 1.6

(*); the mesophilic reference strain at an initial pH 2.0 (^) and pH 1.6 (^); and the

abiotic control at pH 2.0 (�). B: At. ferrooxidans SS3 (initial pH of 2.0 (~) and 2.8 (&))

and the mesophilic reference strain (initial pH of 2.0 (^)) were re-inoculated and

incubated along with an abiotic control (initial pH 2.0 (�)). C. Semi-log plot of iron

oxidation by At. ferrooxidans SS3 using data in Figure 2A and B. Results in the

presence of At. ferrooxidans SS3 and the reference strain are means (n¼ 2)�SD with

single abiotic controls.

Figure 3. Changes in the concentration of (A) dissolved ferrous iron,

(B) dissolved ferric iron, and (C) total dissolved iron during the oxidation of ferrous

sulfate by At. ferrooxidans SS3 at various initial pH values. Symbols: *, pH 1.8; *,

pH 1.9;~, pH 2.0;~, pH 2.2;&, pH 2.5;&, pH 2.8. The data points represent averages

(n¼ 3)� SD.

for iron oxidation by washed cell suspensions of At.ferrooxidans in the range of 20–358C. Both values wereobtained with mesophilic cultures. The Ea values for ironoxidation were 83–95 kJ�mol�1 for iron oxidizers tested inthe temperature range of 2–188C and 4–288C (Ahonen andTuovinen 1989; Ferroni et al., 1986; Franzmann et al., 2005).These higher values are typical in growth situations thatinvolve complex coupling of temperature-dependentcellular processes.

Table II. Generation times of At. ferrooxidans SS3 at various initial pH

values. All cultures were incubated at 58C.

Initial pH g (h) R2

2.8 43.4 0.999

2.5 43.9 0.996

2.2 47.2 0.998

2.0 46.8 0.996

1.9 50.6 0.998

1.8 57.9 0.996

Figure 4. Changes in pH during the oxidation of ferrous sulfate by strain SS3 at

58C. Symbols: *, pH 1.8; *, pH 1.9; ~, pH 2.0; ~, pH 2.2; &, pH 2.5; &, pH 2.8. Each

datum point represents a mean of triplicate experiments� SD.



Figure 5. Growth of At. ferrooxidans SS3 with tetrathionate as sole energy

source at 58 (&) and 308C (*). Data points are averages of 2–3 replicates�SD. Figure 7. X-ray diffraction patterns for precipitates harvested from

At. ferrooxidans SS3 culture grown at 58C (A) and Smolnik mesophilic culture grown

at 308C (B). The peak positions are indicated in Angstroms. Sw, schwertmannite;

J, jarosite.

XRD Analysis of Precipitates

Fe(III)-precipitates for mineralogical analysis were collectedat the end of the experiment from SS3 and the mesophilicreference strain Smolnik grown at 5 and 308C, respectively.The X-ray diffraction patterns of the samples showed thatthe precipitates were solid solutions of schwertmannite andjarosite (Fig. 7). The 58C sample contained schwertmanniteand lesser amounts of jarosite as compared to the 308Csample where jarosite was the dominant phase. Schwert-mannite (Fe8O8(OH)6SO4) is a relatively poorly crystallinemineral commonly found in acid mine drainage sites(Bigham et al., 1996; Gagliano et al., 2004) as well as incultures of iron-oxidizing bacteria (Wang et al., 2007).

Figure 6. Arrhenius plot for the oxidation of iron measured from oxygen uptake

rates with 135 mM Fe2þ at pH 2.0, in the temperature range of 5–308C. The linear

portion of the plot gives an Ea value of 38 kJ�mol�1.


Jarosite (KFe3(SO4)2(OH)6) is common in iron-oxidizingcultures because most media formulations containsufficient Kþ for jarosite precipitation. Schwertmannite isconverted to jarosite with time, at a higher temperature, orwith an increase in the Kþ concentration (Wang et al., 2006).The relative differences in the X-ray diffraction patterns areconsistent with the corresponding incubation conditions:the low temperature (58C) favored schwertmannite pre-cipitation while the higher temperature (308C) was moreconducive to jarosite formation.

Concluding Remarks

To our knowledge, At. ferrooxidans SS3 is the firstpsychrotolerant iron-oxidizing acidophile isolated andcharacterized in pure culture for oxidation of iron andtetrathionate and its phylogenetic affiliation. Its iron-oxidizing enzymes coupled with electron transport andgrowth may provide an interesting evolutionary model oftemperature acclimation in an extremophile. Temperaturealso had a major role in other biogeochemical transforma-tions of iron in this study. Fe(III)-precipitation associatedwith iron oxidation at 58C yielded predominantly schwert-mannite whereas the mesophilic temperature resulted injarosite formation, indicating temperature-dependent dif-ferences in the equilibration of mineral precipitates. Theacid producing nature of these precipitative reactions wasclearly evident from the changes in pH and total soluble ironover time.

The temperature in boreal environments fluctuatesgreatly between the subzero winter months and thesummer when the mean temperature increases above zero.Specifically in the Norilsk area, diurnal fluctuations in thetemperature during the summer subjects micro-organisms

DOI 10.1002/bit

to transient frosts, and even the highest temperatures donot reach the range of 30–408C that is optimal for typicalmesophiles. This temperature regime suggests that thein situ bacterial oxidation of iron is mediated by psychro-tolerant micro-organisms. With transient frosts, diurnalvariation in in situ iron oxidation rates is very likely, too.The results of this study show that low temperatures do noteliminate biological causes of acid mine drainage problemsin boreal environments. The kinetic data demonstrate thatiron oxidation in boreal bioleaching processes is feasible, afinding important for heap and dump bioleaching opera-tions carried out at extreme latitudes or high altitudes. Thepresence of 16S rRNA gene sequences in SS3, SS5, and SK5that are consistent with previously described phylotypesof iron oxidizers in a Norwegian mine site suggests theseboreal sites may have some commonality in the evolvingacidophilic community composition when microbes aredisturbed by mining processes, perhaps associated with thelong spells of low temperatures.

This study and our preceding article (Dopson et al., 2007)clearly demonstrate that a boreal climate is not a deterrent tomicrobial activity in bioleaching processes at outdoortemperatures. Additionally, during intense oxidation ofsulfidic phases, exothermic reactions produce heat that canmaintain elevated temperatures and even sustain thermo-philes in the interior zones in heaps under snow andice cover (Puhakka et al., 2007). Phylogenetically closelyclustered psychrotolerant iron-oxidizers have now beencultured both from mine water impacted sediments andfrom leach liquor in pilot-scale bioleaching tests, allinvolving boreal source samples. Thus, these two studieshold promise that micro-organisms in boreal environmentscan be practicable in bioleaching applications in spite of lowprevailing temperature ranges.

We thank H. Wang and J.P. Gramp (Department of Microbiology,

Ohio State University, Ohio) for setting up the initial enrichments and

F.S. Jones and J.M. Bigham (School of Environment and Natural

Resources, Ohio State University, Ohio) for XRD analysis and help in

interpretation. We thank Norilsk Nickel Group and Boliden AB for

access to the sampling sites. Sampling in the Norilsk region was made

possible through a European Union grant (INTAS 01-0737) received

from the International Association for the Promotion of Co-operation

with Scientists from the New Independent States of the Former Soviet

Union. Sampling in Kristineberg was supported by a grant from

Georange (Sweden). The work was supported in part by the Slovak

State Project No.51/03R 06042 and Slovak Grant Agency VEGA

Grant No. 2/5033/5. Phylogenetic characterization was funded by

the European Sixth Framework Project ‘‘BioMinE’’ (contract 500329).

References

Ahonen L, Hietanen P, Tuovinen OH. 1990. Temperature relationships of

iron-oxidizing bacteria. In: Karavaiko GI, Rossi G, Avakyan ZA,

editors. Moscow, Russia: Centre for International Projects, USSR

Commission for UNEP. p. 21–28.

Ahonen L, Tuovinen OH. 1990. Kinetics of sulfur oxidation at suboptimal

temperatures. Appl Environ Microbiol 56:560–562.

Ahonen L, Tuovinen OI. 1989. Microbiological oxidation of ferrous iron at

low temperatures. Appl Environ Microbiol 55:312–316.

Alfreider A, Vogt C, Babel W. 2002. Microbial diversity in an in situ reactor

system treating monochlorobenzene contaminated groundwater as

revealed by 16S ribosomal DNA analysis. Syst Appl Microbiol 25:

232–240.

Barasan AH, Tuovinen OH. 1986. An ultraviolet spectrophotometric

method for the determination of pyrite and ferrous iron oxidation

by Thiobacillus ferrooxidans. Appl Microbiol Biotechnol 24:338–341.

Berthelot D, Leduc LG, Ferroni GD. 1993. Temperature studies of iron-

oxidizing autotrophs and acidophilic heterotrophs isolated from ura-

nium mines. Can J Microbiol 39:384–388.

Bigham JM, Schwertmann U, Traina SJ, Winland RL, Wolf D. 1996.

Schwertmannite and the chemical modeling of iron in acid sulfate

waters. Geochim Cosmochim Acta 60:2111–2121.

Breed AW, Dempers CJ, Searby GE, Gardner MN, Rawlings DE, Hansford

GS. 1999. The effect of temperature on the continuous ferrous-iron

oxidation kinetics of a predominantly Leptospirillum ferrooxidans cul-

ture. Biotechnol Bioeng 65:44–53.

DopsonM, Hakala AK, Rahunen N, Kaksonen AH, Lindstrom EB, Puhakka

JA. 2007. Mineral leaching at low temperatures by pure and mixed

cultures of acidophilic microorganisms. Biotechnol Bioeng. In press.

Available online at Wiley InterScience (www.interscience.wiley.com).

DOI: 10.1002/bit.21343.

Dopson M, Lindstrom EB. 1999. Potential role of Thiobacillus caldus in

arsenopyrite bioleaching. Appl Environ Microbiol 65:36–40.

Dopson M, Lindstrom EB. 2004. Analysis of community composition

during moderately thermophilic bioleaching of pyrite, arsenical pyrite

and chalcopyrite. Microb Ecol 48:19–28.

Elberling B. 2003. Disposal of mine tailings in continuous permafrost areas:

environmental aspects and future control strategies. In: Kimble JM,

editor. Cryosols: Permafrost-affected soils. Heidelberg: Springer-

Verlag. p 677–698.

Elberling B, Schippers A, Sand W. 2000. Bacterial and chemical oxidation

of pyritic mine tailings at low temperatures. J Contam Hydrol 41:225–

238.

Ferroni GD, Leduc LG, Todd M. 1986. Isolation and temperature char-

acterization of psychrotrophic strains of Thiobacillus ferrooxidans from

the environment of a uraniummine. J Gen Appl Microbiol 32:169–175.

Franzmann PD, Haddad CM, Hawkes RB, Robertson WJ, Plumb JJ. 2005.

Effects of temperature on the rates of iron and sulfur oxidation by

selected bioleaching Bacteria and Archaea: Application of the Rat-

kowsky equation. Min Eng 18:1304–1314.

Gagliano WB, Brill MR, Bigham JM, Jones FS, Traina SJ. 2004. Chemistry

and mineralogy of sediments in a constructed mine drainage wetland.

Geochim Cosmochim Acta 68:2119–2128.

Hallberg KB, Johnson DB. 2001. Biodiversity of acidophilic prokaryotes.

Adv Appl Microbiol 49:37–84.

Holmstrom H, Salmon UJ, Carlsson E, Paraskev P, Ohlander B. 2001.

Geochemical investigations of sulfide-bearing tailings at Kristineberg,

northern Sweden, a few years after remediation. Sci Tot Environ

273:111–133.

Johnson DB, Rolfe S, Hallberg KB, Iversen E. 2001. Isolation and phylo-

genetic characterization of acidophilic microorganisms indigenous to

acidic drainage waters at an abandoned Norwegian copper mine.

Environ Microbiol 3:630–637.

Karavaiko GI, Turova TP, Kondrat’eva TF, Lysenko AM, Kolganova TV,

Ageeva SN, Muntyan LN, Pivovarova TA. 2003. Phylogenetic hetero-

geneity of the species Acidithiobacillus ferrooxidans. Int J System Evol

Microbiol 53:113–119.

Karnachuk OV, Pimenov NV, Yusupov SK, Frank YA, Kaksonen AH,

Puhakka JA, Ivanov MV, Lindstrom EB, Tuovinen OH. 2005. Sulfate

reduction potential in sediments in the Norilsk mining area, northern

Siberia. Geomicrobiol J 22:11–25.

Kovalenko TV, Karavaiko GI, Piskunov VP. 1981. Effect of Fe3þ ions in the

oxidation of ferrous iron by Thiobacillus ferrooxidans at various tem-

peratures. Mikrobiologiya 51:156–160.

Kupka D. 2004. Bacterial oxidation and reduction of iron in the process of

creation and treatment of acid mine waters. Acta Montan Slov 9:467–

472.



Lane DJ. 1985. 16S/23S sequencing. In: Stackebrandt E, Goodfellow M,

editors. Nucleic acid techniques in bacterial systematics. New York:

John Wiley. p 115–176.

Langdahl BR, Ingvorsen K. 1997. Temperature characteristics of bacterial

iron solubilisation and 14C assimilation in naturally exposed sulfide ore

material at Citronen Fjord, North Greenland (838N). FEMS Microbiol

Ecol 23:275–283.

Leduc LG, Trevors JT, Ferroni GD. 1993. Thermal characterization of

different isolates of Thiobacillus ferrooxidans. FEMS Microbiol Lett

108:189–194.

Messing J. 1983. New M13 vectors for cloning. Methods Enzymol 101:

20–78.

Muyzer G, Smalla K. 1998. Application of denaturing gradient gel electro-

phoresis (DGGE) and temperature gradient gel electrophoresis (TGGE)

in microbial ecology. Antonie van Leeuwenhoek 73:127–141.

Muyzer G, Teske A,Wirsen CO, Jannasch HW. 1995. Phylogenetic relation-

ships of Thiomicrospira species and their identification in deep-sea

hydrothermal vent samples by denaturing gradient gel electrophoresis

of 16S rDNA fragments. Arch Microbiol 164:165–172.

Nemati N, Webb C. 1997. A kinetic model for biological oxidation of

ferrous iron by Thiobacillus ferrooxidans. Biotechnol Bioeng 53:478–

486.

Norris PR. 2007. Acidophile diversity in mineral sulfide oxidation. In:

Rawlings DE, Johnson DB, editors. Biomining. Berlin: Springer-

Verlag. p 199–235.

Okereke A, Stevens SE. 1991. Kinetics of iron oxidation by Thiobacillus

ferrooxidans. Appl Environ Microbiol 57:1052–1056.

Olson GJ, Brierley JA, Brierley CL. 2003. Bioleaching review part B. Progress

in bioleaching: applications of microbial processes by the minerals

industries. Appl Microbiol Biotechnol 63:249–257.


Plumb JJ, Hawkes RB, Franzmann PD. 2007. The microbiology of mod-

erately thermophilic and transiently thermophilic ore heaps. In: Rawl-

ings DE, Johnson DB, editors. Biomining. Berlin: Springer-Verlag.

p 217–235.

Puhakka JA, Kaksonen AH, Riekkola-Vanhanen M. 2007. Heap leaching of

black schist. In: Rawlings DE, Johnson DB, editors. Biomining. Berlin:

Springer-Verlag. p 139–151.

Rawlings DE. 2002. Heavy metal mining using microbes. Annu Rev

Microbiol 56:65–91.

Rohwerder T, Gehrke T, Kinzler K, Sand W. 2003. Bioleaching review part

A. Progress in bioleaching: fundamentals and mechanisms of bacterial

metal sulfide oxidation. Appl Microbiol Biotechnol 63:239–248.

Savchenko VA, Novitsky MA. 2001. The climate of Norilsk. Moscow,

Russia: SIP RIA [in Russian].

Silverman MP, Lundgren DG. 1959. Studies on the chemoautotrophic iron

bacterium Ferrobacillus ferrooxidans. I. An improved medium and a

harvesting procedure for securing high cell yields. J Bacteriol 77:642–

647.

Strunk O, Ludwig W. 2002. ARB-a software environment for sequence

data: Department of Microbiology, Technical University of Munich,

Germany

Vogel AI. 1961. A text-book of quantitative inorganic analysis including

elementary instrumental analysis. New York: John Wiley and Sons.

Wang H, Bigham JM, Jones FS, Tuovinen OH. 2007. Synthesis and proper-

ties of ammoniojarosites prepared with iron-oxidizing acidophilic

microorganisms at 22–658C. Geochim Cosmochim Acta 71:155–164.

Wang H, Bigham JM, Tuovinen OH. 2006. Formation of schwertmannite

and its transformation to jarosite in the presence of acidophilic iron-

oxidizing microorganisms. Mater Sci Engin Part C Biomim Supramol

Syst 26:588–595.

DOI 10.1002/bit

Documents

Bacterial oxidation of ferrous iron at low temperatures